Details of the Relation between XML and RDBMS - Term Paper Example

Table of contents Introduction: 2 Background: 2 Is XML and databases are really complementary 5 XML advantages over RDBMS: 5 Data Integrity: 6 Instances where XML is not a complimentary: 9 When to consider other Databases 10 Barriers to XML to treat as DBMS: 11 Introduction: Today many software applications need integration of small and large data from various sources. People are very conservative about their database choices. The flow of information is just as important as the control of information. One of such forms of integration is XML with RDBMS. Though both the technologies have their own advantages, disadvantages and interoperability, there are some instances where one is proffered more than the other. Hence the current study makes an insight into the minute details of the relation between XML and RDBMS to understand their true nature of complimentary and controversy. Background: Basically, XML document is a collection of data. Though by collection of data, XML document is said as database in other sense a mere collection of data doesn't make much difference between the other files that store data. After all, all files contain data of some sort. Still as a "database" format, XML has some advantages than other forms of data collection. XML in a database format is a self describing, portable, and can describe data in tree or graph structure. It is self-describing in the sense that the markup describes the structure and type names of the data, although not the semantics, it is portable means it uses Unicode, and it can describe data in tree or graph structures. XML is treated as database because of its surrounding capabilities like: storage (XML documents), schemas (DTDs, XML Schemas, RELAX NG, and so on), query languages (XQuery, XPath, XQL, XML-QL, QUILT, etc.), programming interfaces (SAX, DOM, JDOM), like other databases. Hanson expresses that "XML data are extremely well-suited to hierarchical storage," But long winded nature and the slow accessibility to data due to parsing and text conversion are the drawbacks of this technology. Also on the minus side, it lacks many of the things found in real databases: efficient storage, indexes, security, transactions and data integrity, multi-user access, triggers, queries across multiple documents like RDBMS. Reasons to adopt XML database: According to John Cox, senior editor of Network World, there is need for the XML database as the network executives begin to experiment with web services. Users will be interacting with business partners and customers through web services. On the other hand, suppose Web site is built from a number of prose-oriented XML documents. Not only to manage the site, also to provide a way for users to search its contents. Documents are likely to have a less regular structure and things such as entity usage are probably important because they are a fundamental part of how documents are structured. In this case, one needs a product like a native XML database or a content management system, which allow to preserve physical document structure, support document-level transactions, and execute queries in an XML query language. And in case of object oriented applications, one needs a system that can store those objects in the database or serialize them as XML. A good example of the type of "database" for which an XML document is suitable is an .ini file -- that is, a file that contains application configuration information. It is much easier to invent a small XML language and write a SAX application for interpreting that language than it is to write a parser for comma-delimited files. In addition, XML allows you to have nested entries, something that is harder to do in comma-delimited files. However, this is hardly a database, since it is read and written linearly, and then only when the application is started and ended. XML in web development: XML is really helpful when sharing or transferring data between web and mobile, for some applications a database is a much better idea than XML, but not all. XML is good for storing text, especially complex, hierarchically-related text, because that's what XML designed for. It's also turned out to be fairly good for data transfer between disparate systems. But it would not be useful for to store several billion customer account records for a real time banking operation like an ATM system, for example, although it might be quite good for the messaging between component sites. Examples database for which an XML document might be suitable as a database are personal contact lists (names, phone numbers, addresses, etc.), browser bookmarks, However, given the low price and ease of use of databases, there seems little reason to use an XML document as a database. The only real advantage of XML is that the data is portable, and this is less of an advantage than it seems due to the widespread availability of tools for serializing databases as XML. Is XML and databases are really complementary Ronald Bourret while explaining about the relation between XML and databases classifies XML documents in two ways - Data centric documents and document centric documents. In data centric documents, xml is used as data transport, where attributes and entities are not important in data centric documents. Applicable areas include scientific data, sales data and dynamic WebPages where the functional data architecture is much important than the physical structure. Where as in document centric documents are those user manuals, static WebPages, those are with irregular structure and mixed content where the physical structure is important For the purpose of storing and retrieving the data from those documents, some data transfer is needed, this can be build with database RDBMS. XML advantages over RDBMS: There are several advantages of XML database compared to RDBMS: simplicity, ease of application development, fast document retrieval. Xml database stores and manages the increasing number of XML documents efficiently. There is also better support from vendors to XML. Microsoft's Yukon release of SQL Server and Oracle's Project XDB demonstrates such support which treats XML documents as a new data type and manage them to work efficiently with relational data and objects. XML as RDBMS: If there is relational database architecture using XML as a backend, one xml file refers to permissions applied to a specific action (like user/add/user or user/ edit/ document). Another xml file refers to roles. So, each role has permissions for sub nodes it contains. Finally another xml file refers to all the uses which have a mixture of roles and specially assigned permissions for specific resources. The whole solution takes less than 2kb, and the code was only about 100 lines long that specified whether a user had access to a specific resource. Jonathan Robie said that Whether physical XML or a virtual view, data integration is one of the biggest reasons for XML, and providing database functionality for the XML view is one of the biggest reasons for an XML database. Another real power of XML is in hierarchical data. Since projects can have sub-projects, addition of child node for each sub-project requires a little more complicated process in a typical relational database. Whereas in XML, as the data is hierarchical data XML has more advantage which stores relatively small amount of data in memory. Instances of complimentary: Data Integrity: Referential integrity is core of the relational database model, but xml notion of integrity is not clear. Michael Champion explains that If anyone want to change the items information in XML representation (e.g., the customer address in a purchase order should change when the address in the customer database changes), RDBMS-based solutions are generally more appropriate. XML has no good notion of "referential integrity" and that could bite hard. The real benefits of XML are not aimed at internal systems, such as reading own data from a database on own web server, and then displaying it from PHP. XML Sole Advantages: XML is a really convenient way to get two or more different systems (computers, software, data management tiers) to agree with each other on how to transmit data. For example, if using the mysql, while transferring the data to another machine and to recreate the database, data has to be converted into text file to transfer to another machine. Butt it is good when using the mysql only. But while placing that data in PDF document or a windows help system file, there is a need of XML, to transfer the data to another medium. This is because of the neutral format of the data that can be analyzed by many different types of software. This is only one example of a good use of XML. GUI NAVIAGATION It's also worth noting that XML allows the developer to store data in a tree format, where as traditional databases do not. Also this gives GUI navigation. It would be a great way to document all the files and directories on a computer, since that information is in a tree format. It is also a good way to transmit information about a GUI navigation or menu system. Data Insertion: To insert a new XML tag, say "OrderNo", as another property of "Customer Address", one can easily run the following Query: INSERT INTO xml table VALUES ('Order No','Cust Address','Customer',NULL); The entity will be inserted. Here as the ancestors' addresses are already present in the pointers of the entity, there is no necessity to update the parent node while inserting a node into an XML tree. Also in order to add a successor node, a parent's field need not be updated Competing factors: i). Storage space: A data administrator, Larry Hanson pointed out that XML data is fundamentally different from relational data. "In XML databases, an online tax return can be stored in its entirety. In a relational database, each line of the return would have to be a different table of data in rows and columns." Trying to "force fit" an XML document into the rigid relational structure can waste storage space and lead to inefficiencies in queries and retrievals. ii). Efficiency: Object-oriented databases provide storage of objects but lack fast access. Whereas Xml through its hierarchical structure of data storage with fair linear performance offering retrieval efficiency. But it uses a bit of extra space for storing successor links. XML's best feature is its efficiency in operations to store and retrieve XML data by using only one-run SQL query, where as representing XML data in terms of the rows and columns of a relational table is quite inefficient (A. Eli, B. Rahnama) Instances where XML is not a complimentary: i). The situation of tagging content with XML when using a database is the one such case of interoperability. Say for example, column named "title" in database table can be written in PHP script to display "title" from the database. But when it is written in with XML tag such as "" and save that along with the title data in the database, it will become "title" in database. In such instances XML has absolutely no benefit. ii). XML Vs databases: The xml is not at all useful at all for dynamic data driven web site, where the information on web site changes regularly .The data is to be in database is the best when queries have to operate on the data .For example: to count the articles published on website. Use of databases seems to be harder and slower. In this case it is good t o use XML to exchanging the data between different systems. Database is a great tool. The facility to use the sql queries to get the data makes the database as a powerful tool. But on the other hand, distributing an application which uses a database can be a great pain. The application which uses xml as a database, allows one to scale searches on great amount of data with little overhead, but the disadvantage is Xml data is not going give all the tools as database will. On the other hand distributing of XML is much easier than distributing database. Considering a layer of abstraction between application and the data source is a good thing while determining the type access needed to get the data ahead of time. The problem is that, with a database, one can do a lot more than with the XML file - for example if searches begin with using "STARTS WITH" are perfectly fine with some databases but are going to be a tolerate with an XML file. Hence, creating one single abstraction layer might involve difficult for both an XML backend and a DB backend. iii). While using xml files to store content management data vs. a database, it would be difficult to implement a site-wide search function using xml. It would need some kind of index file to define the files to be search and also it would involve lots of IO on the server. It is good to use MySQL if searching through content is needed. The best example for XML database is JStore. JStore is an underlaying component of JALIOS.JCMS is a French content management system. The power of Jstore as a database it can hold up to 80000 A4pages, store loads from some seconds to few minutes. CMS needs to change object structure, hierarchy, pointer quickly without thinking about database structure. So, it was designed using XML. When to consider other Databases In the applications like e-commerce apps where there is heavy involvement of transactions and queries, there is a tendency to think about highly structured data. In such cases, the developers keeps interest on how to store data and not on how it is stored in XML, entities and the encodings used by xml documents is not important in this case. Here comes the necessity for a relational database and needs to transfer the data between XML documents and the database, if at all an XML is also used. When the data consists thousands of rows and columns RDBMS is best solution. Barriers to XML to treat as DBMS: Existing RDBMS technology have more performance than XML. The good speed of data retrieval through RDBMS in most data scenarios and also a group of related technologies riding on the back of XML failed to give an equal status to XML with RDBMS. Data administrators are not encouraging the communication between XML and database Recommenders of databases, who allows to access information in relational databases can easily point out the difficulties of XML in distributed data. But some of the authors still agree with the performance of both technologies. For example Jonathan Robie acts as a diplomat between two disparate technologies- XML and RDBMS. But the task of communication between these two technologies is not a task for the faint-hearted. (it is yet to be agree though some companies following). Document oriented xml has a varying structure to allow the flexibility in text. But all the semi structure data is not document oriented. It is possible to record the semi structured data into relational data base model. But there will be performance overheads particularly in querying. So the relational databases are good to store highly structured data Rigorous Structure: While storing the Xml schema with a fixed structure which has no mixed content or recursive content model, it is good to use RDBMS in this case. Michael Champion explains that in database technology, information is to be queried, RDBMS is the better option than XML. When there is need to manipulate a recursive content model, then XML query language will be better than RDBMS. For example, XML query language is useful than RDBMS to know the products which contains assemblies or sub assemblies or sub-sub-assemblies. The RDBMS approach is useful than XML when there are queries which involve joins across metadata in different collections Reasons to consider the complementary approach: However the hybrid approach model that is the use of XML and RDBMS together solves all the data problems in the following ways: XML is good for represent hierarchical data (tree structure of data), where as RDBMS is good at managing data oriented data. Also to interact with OS file, hierarchical data model is used The hybrid approach fulfills the requirements of long term data storage Applications that needs to query subset of data uses RDBMS and XML is for distribution of the data as an effective technique. Shared Repository: If any one wants to add an xml view onto exiting database or repository, there is need to consider the benefits of enterprise integration of native XML database and also data integrity issues. It's also a necessary to share the data with applications which are not XML-aware; this is often true for data-oriented XML. In these situations an RDBMS or XML enabled Database gives the best. For many situations, it can be achieved by relational database, especially if the majority of the applications representing from that database may not need to manipulate or view the data as XML. In this case, XML Middleware, or the mapping facilities that RDBMSes are providing directly is sufficient (Tom Bradford) Ronald Bourret says that the native xml database does not return the data in non XML databases but the data in an XML-enabled relational database is accessible to both XML and non-XML applications. Enterprise Integration: In data oriented applications, Xml can be used for integration or data exchange between enterprise applications inside and outside of the firewall, Michael Champion says that XML DBMSes are solutions for all kind of problems, especially in the area where XML is the "lingua franca" of electronic commerce. There is a possibility of normalization of XML purchase orders which come from SOAP and Web pages into 20-30 RDBMS tables, with the help of an army of programmers and allows DBAs to manage with diversity and evolution and also it is simpler with XML DBMS. Conclusion: Native XML databases looks like a good way to integrate data from a variety of backend, and in future, native XML databases will do the data integration transparently and bi-directionally, where as Relational databases have problems in this area. Since the result of integrating data from a variety of backend sources is likely to be a semi-structured data, not structured data, Integrating of data from a variety of backends is one of the promising areas for native XML DBMS, and that ability to handle semi-structured data is a key benefit. Xml can be used as a database in the areas of requirement like small amount of data, few users and modest performance, but it is not in the case of requirements of strict data integrity, many users and need of good performance XML and relational databases are really complementary. XML is not really a storage format. It is more a serialization format. It is a convenient format to pull data from an application or a relational database and send it in a machine independent manner to another machine. While database is excelling in storing/retrieving data, XML excels in representing an interchange format for the data. So, XML is not a replacement for a database, rather is a completely different concept from databases, but can be used well in conjunction with RDBMS. Reference: A Bite of XML: XML andDatabase, August 19, 2006, accessed on 12 July 2007, available online from http://abiteofxml.wordpress.com/2006/08/19/xml-and-databases/ Devx discussions: DevX.com Forums > DevX Developer Forums > XML >XML files vs Database, available online from http://forums.devx.com/showthread.phpt=4567 Joab Jackson, SQL vs. XML in a database world, accessed on 12 July 2007, available online from< http://www.gcn.com/print/24_6/35285-1.htm> Fabian Pascal, The XML Bug , accessed on 13 July 2007, Available online from Rosen, 2007, 'Xml as a RDBMS', discussion group, 'database development', accessed on from 13 July 2007, available online from Leigh Dodds, XML and Databases, October 24, 2001, accessed on 12 July 2007, available online from http://www.xml.com/pub/a/2001/10/24/follow-yr-nose.html Dan Zambonini,The difference between XML and RDF, September 20, 2005, accessed on 12 July 2007, available from Dare Obasanjo, AN EXPLORATION OFXML IN DATABASE MANAGEMENT SYSTEMS, accessed on 12 July 2007, available online from Working out the bugs in XML databases, accessed on 12 July 2007, available from < http://www.networkworld.com/news/2002/0107specialfocus.html> XMl files vs. Database, accessed on 12 July 2007, available from < http://www.codeguru.com/forum/archive/index.php/t-204260.html> Database vs. XML: Dev Articles Community Forums >Databases >Database Development , accessed on 12 July 2007, available online from < http://forums.devarticles.com/database-development-6/database-vs-xml-273.html> Is XML a database accessed on 12 July 2007, available online from Is XML a database accessed on 12 July 2007, available online from Database VS XML, accessed on 12 July 2007, available online from Ronald Bourret, XML and databases, accessed on 12 July 2007, available online from Read More